Does plant diversity influence phosphorus cycling in experimental grasslands?

Geoderma 167-168 (2011) 178–187

Contents lists available at SciVerse ScienceDirect

Geoderma journal homepage: www.elsevier.com/locate/geoderma

Does plant diversity influence phosphorus cycling in experimental grasslands? Yvonne Oelmann a,⁎, Anika K. Richter b, Christiane Roscher c, Stephan Rosenkranz b, Vicky M. Temperton d, Wolfgang W. Weisser e, Wolfgang Wilcke f a

Geoecology/Geography, Eberhard Karls University of Tübingen, Rümelinstr. 19-23, 72070 Tübingen, Germany Geographic Institute, Johannes Gutenberg University Mainz, Johann-Joachim-Becher-Weg 21, 55128 Mainz, Germany Helmholtz Centre for Environmental Research – UFZ, Community Ecology, Theodor-Lieser-Strasse 4, 06120 Halle, Germany d Phytosphere Institute ICG3, Jülich Research Centre GmbH, Leo-Brandt-Strasse, 52428 Jülich, Germany e Institute of Ecology, Friedrich Schiller University of Jena, Dornburger Straße 159, 07743 Jena, Germany f Geographic Institute, University of Berne, Hallerstrasse 12, 3012 Berne, Switzerland b c

a r t i c l e

i n f o

Article history: Received 25 February 2011 Received in revised form 16 August 2011 Accepted 18 September 2011 Available online 2 November 2011 Keywords: Phosphorus Plant diversity Plant P uptake P fractions in soil P in soil solution

a b s t r a c t Plant diversity was shown to influence the N cycle, but plant diversity effects on other nutrients remain unclear. We tested whether plant species richness or the presence/absence of particular functional plant groups influences P partitioning among differently extractable pools in soil, P concentrations in soil solution, and exploitation of P resources (i.e. the proportion of total bioavailable P in plants and soil that was stored in aboveground biomass) by the plant community in a 5-year biodiversity experiment in grassland. The experimental grassland site established in 2002 had 82 plots with different combinations of numbers of species (1, 2, 4, 8, 16, 60) and functional groups (grasses, small non-leguminous herbs, tall non-leguminous herbs, legumes). In 2007, we determined P partitioning (Hedley) in soil of all experimental plots. We sampled plant community biomass and continuously extracted soil solution with suction plates from March 2003 to February 2007 and determined PO4-P concentrations in all samples. The presence of legumes increased aboveground P storage in plants and decreased labile Pi concentrations in soil because of their higher demands for P associated with N2 fixation. During cold periods, readily plantavailable PO4-P concentrations in soil solution increased in legume-containing mixtures likely caused by leaching from P-rich residues. We found a consistently positive effect of plant species richness on P exploitation by the plant community which was independent of the presence of particular plant functional groups. With proceeding time after establishment, plant species richness increasingly contributed to the explanation of the variance in P exploitation. Therefore, plant strategies to efficiently acquire P seem to become increasingly important in these grasslands. We conclude that diverse plant communities are better prepared than less diverse mixtures to respond to P limitation induced by continuously high atmospheric N deposition. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Plant species richness is partly driven by nutrient supply (Janssens et al., 1998; McCrea et al., 2001; Wassen et al., 2005). However, at similar nutrient supply, plant species richness in turn has been shown to influence ecosystem functioning including the nutrient cycle (Scherer-Lorenzen et al., 2003; Spehn et al., 2005; Tilman et al., 1996). A number of manipulative field experiments have been conducted to investigate the effect of plant species richness on ecosystem processes to assess the potential consequences of the currently observed biodiversity loss (Hooper et al., 2005; Schmid et al., 2002). Most experimental studies have focused on biomass production and the N cycle so far, since N limits plant productivity at many

⁎ Corresponding author. Tel.: + 49 70712972398; fax: + 49 7071295378. E-mail address: [email protected] (Y. Oelmann). 0016-7061/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.geoderma.2011.09.012

locations of the temperate zone (Vitousek and Howarth, 1991). As a result of anthropogenic N deposition and the associated shift in N:P ratios in biomass, however, P is becoming increasingly important for plant growth and biodiversity (Galloway et al., 2004; Vitousek et al., 2010; Wassen et al., 2005). The underlying mechanistic hypothesis to explain the plant species richness-nutrient cycling relationship is that plants can use available nutrients in a complementary way (Trenbath, 1974). Thus, niche differentiation in space and/or time in more diverse systems may result in a more complete resource use at the community level compared to less diverse systems (Hooper et al., 2005). Positive complementarity effects have explained increased biomass production in a number of plant diversity experiments (Marquard et al., 2009; Spehn et al., 2005; van Ruijven and Berendse, 2003). If plants take up more N because of increased biomass production in diverse systems, the uptake of all other essential elements, e.g. P, must also increase. Karanika et al. (2007) presented support for this hypothesis based on a pot

Y. Oelmann et al. / Geoderma 167-168 (2011) 178–187

experiment. It remains, however, unclear whether any positive relationship between plant diversity and P exploitation holds true for established grassland systems. Driven by complementary N uptake, increasing plant species diversity decreased plant-available soil N concentrations (Niklaus et al., 2001; Scherer-Lorenzen et al., 2003; Tilman et al., 1996). This effect was also observed at our study site, the Jena Experiment (Oelmann et al., 2007b) where, however, the effect of plant species diversity on the P cycle has not yet been tested. Hooper and Vitousek (1998) found an effect of functional groups on resin-extractable P in soil mainly caused by the difference between vegetated and non-vegetated plots. The study of the plant diversity-P cycle relationship is in contrast to N complicated by the strong physicochemical control of P solubility (Hinsinger, 2001). Furthermore, it is difficult to identify the plantavailable P fraction in soil, because i) extractions with salt solutions often yield no correlations between P concentrations in the extraction solution and measures of plant P uptake (Gilbert et al., 2009), and ii) the soil solution as the most sensitive measure of P availability is characterized by low P concentrations due to the low solubility product of the associated P-containing minerals posing problems to analytical accuracy. Therefore, several P fractions ranging from the instantaneous measure of P concentrations in soil solution to total P storage in soil must be included to gain insight into the relationship between plant diversity and P availability in soil. Although P concentrations in either soil (labile fraction) or plant material is commonly used for evaluating bioavailability of P (Alt et al., 2011; Critchley et al., 2002; Gilbert et al., 2009; Janssens et al., 1998), to our knowledge, no study has linked aboveground P exploitation and P fractions in soil of differently diverse grasslands yet. The potential effect of plant diversity could vary among different soil depths and among seasons because of complementary P uptake in space and time as was observed for N (Bardgett et al., 2003; McKane et al., 2002; von Felten et al., 2009). Diversity in functional plant properties relevant for nutrient acquisition and use among co-existing plant species might favor complementary P uptake. Therefore, the consideration of plant functional groups for P cycling in differently diverse grassland systems is essential. For example, the high P demand of legumes because of the energy costs associated with N2 fixation which is supplied by the synthesis of adenosinetriphosphate (ATP) is well known (Aerts and Chapin, 2000; Pate, 1986). Because of their extensive rooting system, grasses might more efficiently explore P resources in soil and thus, reduce PO4-P concentrations in soil solution (Hooda et al., 1999). However, the exploitation of nutrient resources in soil by grasses might not be as efficient for P as was reported for N, because of reduced infestation of grasses with arbuscular mycorrhizal fungi (Brundrett, 2002). All reported plant diversity or functional group effects on above- and belowground nutrient pools in grasslands were based on short-term experiments. In short-term studies weather conditions might strongly affect biomass production (Ciais et al., 2005) and thus, short-term ecosystem P cycling. Therefore, plant diversity or functional group effects might be restricted to a particular study year. Furthermore, in particular belowground processes are known to respond to plant diversity only after a time lag (Eisenhauer et al., 2010). Recently, Oelmann et al. (2011) have shown that the relationship between plant diversity and N availability in soil changed systematically during the first five years after establishment of the experimental grassland mainly induced by organic matter accumulation because of conversion from arable land to grassland. Availability of P in soil might change with time because of periodic removal of nutrients with the mowing and harvesting of grassland plants and continuous depletion of P-fertilizer remains (applied during agricultural use frequently preceding conversion to grassland in central Europe). Therefore, it is likely that competitive interactions among plant species – mainly driven by resource availability – are changing and reduced P availability with time also modulates plant diversity effects. Taken together, these arguments illustrate that

179

longer-term studies are required to derive general conclusions about plant diversity or functional group effects on P cycling. The objective of our study was to test if plant species richness per se or the presence of plant functional groups influences P cycling i.e., P exploitation by the plant community, P concentrations and partitioning in solid soil, and P concentrations in soil solution in artificial grassland systems established by conversion from arable land during the first five years after sowing. We hypothesized that (i) plant species richness is positively correlated with P exploitation of plant communities and correspondingly negatively with P availability in soil similar to N, (ii) plant functional identity influences aboveground P exploitation of plant communities and P availability in soil, and (iii) plant species richness gains importance for P exploitation of plant communities and P availability in soil with time after establishment of the grassland. 2. Materials and methods This study was conducted in the framework of the “Jena Experiment” (www.the-jena-experiment.de) which addresses the role of biodiversity on element cycling and trophic interactions in experimental grassland systems (Roscher et al., 2004). 2.1. Study site The field site is located near the German city of Jena (50°55′ N, 11°35′ E; 130 m above sea level). Mean annual air temperature is 9.3 °C, and mean annual precipitation amounts to 587 mm (Kluge and Müller-Westermeier, 2000). A comparison of this long-term mean (LTM, 1960–1990) with the study years is given in Table 1. The soil is an Eutric Fluvisol developed from up to 2 m-thick fluvial sediments that are almost free of stones. The systematic variation in soil texture as a consequence of fluvial dynamics is considered in the experimental design by arranging the plots in four blocks parallel to the river. The study site was converted from grassland to arable land in the early 1960s. Organic C concentrations ranged from 13 to 33 g kg −1, organic C:N ratios from 8 to 15, and pH (H2O) from 7.1 to 8.4 at the start of the experiment in 2002. The experimental design is described in Roscher et al. (2004). Briefly, the main experiment comprises 82 plots (20 × 20 m) of different levels of species richness (1, 2, 4, 8, 16, 60) and different numbers (1, 2, 3, 4) of plant functional groups (grasses, [non-leguminous] small herbs, [non-leguminous] tall herbs, legumes) chosen from a species pool of 60 species representing species typically occurring in Molinio-Arrhenatheretea meadows (Ellenberg, 1996). Each plantdiversity level had 16 replicates except for 14 mixtures with 16 species and four replicates of the 60-species mixture. The management of all plots was adapted to extensive meadows used for hay production, and all plots were mown twice a year in June and September. Plots were not fertilized during the experimental period. To maintain the sown species diversity level, plots were weeded regularly by cutting the weeds at the soil surface. Table 1 Weather conditions during the study period. Significant differences between the LongTerm Mean (LTM) and the respective years are indicated by different letters. For statistical details see methods. no. = number, n.a. = not available

LTM 2003 2004 2005 2006 2007 a

Mean T (°C)

Rainfall sum (mm)

9.3 a 9.9 a 9.4 a 9.2 a 10.0 a 10.2 a

587 436 573 422 493 710

no. = number.

a b a b a a

No.a of days T b 0 °C

No.a of days T b 0 °C

(March)

(Winter)

n.a 0 1 8 14 0

n.a 30 37 47 6 n.a

180

Y. Oelmann et al. / Geoderma 167-168 (2011) 178–187

2.2. Sampling and chemical analyses In May and August from 2003 to 2007 – at estimated peak biomass before mowing, aboveground plant biomass was harvested on all plots within a frame (0.2 × 0.5 m, height 0.03 m) at four random locations per plot, sorted to species level and dried at 70 °C for 48 h. Dried plant material from living tissues of all subsamples per plot and per harvest campaign were pooled and ground with a Cyclotec 1093 Sample Mill (Foss Tecator, Hoganas, Sweden) using a 0.5 mm screen for chemical analysis. To determine P concentrations in aboveground biomass, each sample was digested with HNO3 at 260 °C and at about 600–700 MPa using the microwave-assisted high pressure digestion unit “Multiwave” (Perkin-Elmer, Rodgau-Jügesheim, Germany). Unfortunately, samples of fall 2003 got lost during mail delivery. To get an estimate of P storage for the missing samples, we used the mean fall to spring ratio of all study years of aboveground P storage per plot and calculated P storage in fall 2003 based on data of the spring harvest 2003. Phosphorus partitioning in soil was assessed in September 2007 (five replicates, 1 cm diameter, bulked to one composite sample per plot). We used a four-step sequential P fractionation after Hedley et al. (1982) modified by Kuo (1996). Sequentially, we used 20 ml NaHCO3 (adjusted to pH 8.5), 30 ml NaOH, and 35 ml HCl as extraction solutions for 0.5 g soil. The last step comprised the combustion (550 °C) of the remaining soil to destroy all organic material followed by shaking with 20 ml H2SO4. Organic P (Po) concentrations of the respective fractions were calculated as the difference between total dissolved P (TDP) and inorganic P (Pi). Duplicate extractions of P fractions of selected samples yielded the following precision (as standard error of replicate analyses): NaHCO3-Pi 2.2 mg kg −1; NaHCO3-Po 1.6 mg kg −1; NaOH-Pi 5.3 mg kg −1; NaOH-Po 2.7 mg kg −1; and HCl 7.0 mg kg −1. In accordance with Negassa and Leinweber (2009), labile P refers to NaHCO3-Pi and -Po, moderately labile P to NaOH-Pi and -Po and stable P to HCl-P and H2SO4-P fractions. As a measure of readily plant-available P, we also assessed P concentrations in soil solution. Because of financial and work-load restrictions, we used three of the four experimental blocks with a total of 63 plots for the installation of suction plates to collect soil solution. Four months before the collection period, at each of the soil depths of 0.1 and 0.2 21 suction plates (UMS, Munich, Germany, sintered glass, pore size 1 to 1.6 μm) were installed in Block 2 (Oelmann et al., 2007a). At 0.3 m, 63 suction plates were installed in Blocks 1–3 (Oelmann et al., 2007b). Because each block represents a subset of the complete design, no plant species richness or functional group treatments were missing (with the only exception that there was no 16species mixture of only one plant functional group in Block 2). All suction plates were coupled with a permanent vacuum system to collect soil solution. The vacuum system was regulated with the help of manual measurements of soil matric potential. Starting in the second year after seeding, we collected cumulatively extracted soil solution every two weeks from March 2003 to May 2007. To determine seasonal effects we aggregated results of samples of the spring (March to May, 2003 to 2007), and winter periods (December to February, 2003 to 2007). In summer and fall (June to November) dry conditions frequently precluded the extraction of soil solution. Therefore, these periods would not guarantee a consistent data set and were excluded from further analyses. In total, about 4000 soil solution samples were analyzed. Phosphate concentrations in soil solution and soil and plant extracts were measured photometrically (molybdenum blue-reactive P; Murphy and Riley, 1962) with a Continuous Flow Analyzer (CFA; for samples collected from spring 2003 to spring 2004: CFA SAN ++, Skalar [Breda, The Netherlands]; for samples collected between winter 2004/05 and winter 2006/07: CFA Autoanalyzer [Bran&Luebbe, Norderstedt, Germany]). Phosphate concentrations of P fractions in soil in 2007 and of plant digests from 2003 to 2007 were analyzed

using the CFA Autoanalyzer (Bran&Luebbe, Germany). In addition, we analyzed TDP concentrations in the extracts of the respective P fractions by irradiation with UV and oxidation with K2S2O8 followed by reaction with ammonium molybdate described above for PO4-P (Skalar catnr. 503-553w/r). The detection limits were 0.02 mg P l −1 (CFA, Skalar) and 0.04 mg P l −1 (Autoanalyzer, Bran&Luebbe), respectively. Calcium concentrations in soil solution were measured with an AAS 3300 (Perkin Elmer, Rodgau-Jügesheim, Germany). The pH values in soil were determined with a glass electrode (WTW, Weilheim, Germany) in a 1:2.5 soil:water suspension. 2.3. Calculations and statistical analyses To eliminate concentration/dilution effects on readily plantavailable P concentrations in soil solution, we calculated volumeweighted mean (vwm) PO4-P concentrations for each plot for the respective sampling dates using the collected volumes of soil solution as weights. Alternatively, we used arithmetic means which did not change the main results (not shown). We calculated aboveground P storage by multiplying aboveground biomass production and P concentrations in aboveground biomass. We calculated plant community P exploitation (Pe) to eliminate interfering effects of a priori differences in P availability in soil (Eq. [1]): Pe ¼

Pb ×100 Pb þ Pl þ Pm

ð1Þ

based on aboveground P storage (Pb) and plant-availability of P in the intermediate run i.e., the sum of labile (Pl) and moderately labile P (Pm) storage in soil (Cross and Schlesinger, 1995; Hedley et al., 1982) and we assumed no changes in concentrations of the sum of labile and moderately labile P concentrations in soil during the study period. We based our assumption on the following argumentation i) labile P is variable in time (Styles and Coxon, 2007) but represents only less than 7% of total P in our study, ii) the stability of moderately labile P is controversially discussed (Cross & Schlesinger, 1995), but Negassa and Leinweber (2009) reported that in non-fertilized grasslands even 19 years of management did influence P fractions only negligibly. Unfortunately, we do not have information on concentrations of labile and moderately labile P concentrations in soil before establishment of our experiment. However, for a subset comprising seven plots P concentrations of all P fractions are available for 2002, the year in which the plant mixtures were seeded (and thus, the plant influence can be considered as minor). Comparing these data with the respective results of 2007, we found no significant differences in any P fraction or the sum of all fractions (matched pairs t test, pN 0.2). Phosphorus storage of the two fractions in soil was calculated for a soil depth of 0.2 m by considering the mean soil density of each block (values in kg m−3, Block 1: 1270, Block 2: 1170, Block 3: 1240, Block 4: 1250 determined in 2002 with the help of a 100 cm3 steel core). For statistical analyses, we used log-transformed data to improve homogeneity of variances except for pH, proportions of P fractions (already homogeneous variances), and PO4-P concentrations in soil solution (because of concentrations below one conversion to log would increase variation in data sets). Type I sums of squares were applied and the order of fitting or a factor in the model was altered to assess relative strengths of different factors. For all statistical analyses, we used the SPSS software package (SPSS 11.5, SPSS Inc., Chicago, IL, USA). To test for temporal consistency of plant species richness effects (aboveground P storage, P exploitation), a repeated measure ANOVA was performed with plant species richness, presence/absence of legumes, grasses, small herbs, and tall herbs as between-subject factors and time as the within-subject factor (GLM, type I sum of squares). For each functional group, the effect of the presence or

Y. Oelmann et al. / Geoderma 167-168 (2011) 178–187

3. Results

a 10

A

A

P exploitation(%)

A

B

8 C

6 4 2 0

b 12

absent

Aa

present

10

Ba

Ba

Ba

Ba

N:P ratio

absence was tested in separate models (each including block, plant species richness, and functional group richness as first factors). To ease reading, the results of these separate analyses were combined in one table. We used the Greenhouse–Geisser correction to adjust significance values if the assumption of sphericity was not met. Separate ANOVAs per year were performed using the same model as discussed above (time excluded) to elucidate effects of the plant community composition on aboveground P storage, P exploitation, and readily plant-available P concentrations in 2003 to 2007, and on P fractions in soil in 2007. In case of significant plant species richness effects, we additionally ran models using reversed order with respect to plant functional group presence/absence and plant species richness (i.e., Block, functional group presence/absence, plant species richness, functional group richness). This procedure enables us to derive the significance and proportion of explained variance of plant species richness that remains in addition to any functional group effect that was fitted before (hierarchical approach). Post-hoc tests were performed to assess differences among plant species richness levels or functional groups (Tukey in case of homoscedasticity and Games Howell in case of heteroscedasticity). With readily plant-available PO4-P concentrations in soil solution, a matched paired t test indicated significant differences among soil depths and subsequent spring/winter periods. The same test was used for differences in weather conditions including monthly values as matched replicates (Table 1). We are aware that the replicates are not independent. However, the random exclusion of matched pairs did not change results. Therefore, we consider the analyses robust and not influenced by dependency.

181

8

Ab

Bb

6

Cb

Bb

Db

4 2 0 2003

2004

2005

2006

2007

Fig. 1. Temporal course of (a) plant P exploitation, and (b) N:P ratios in aboveground biomass from 2003 to 2007. Different upper case and lower case letters refer to significant differences among years and mixtures with and without legumes, respectively.

3.1. Community biomass, P concentrations, and P storage Analyses of aboveground biomass production of the Jena Experiment from 2003 to 2007 have been published in Marquard et al. (2009). Phosphorus concentrations in plants and aboveground biomass production are negatively related both for the spring and fall harvests (0.28 b r b 0.70; p b 0.02). In spite of the significant variation in P concentrations of aboveground biomass, aboveground P storage is driven by aboveground biomass production because of six orders of magnitude higher biomass production compared with P concentrations in biomass. From 2003 to 2007, N:P ratios in aboveground biomass ranged from 3 to 18 and from 3 to 12 in mixtures with and without legumes, respectively. During this period, we found N:P ratios above 16 in 0 to 5% of legumecontaining mixtures. In mixtures without legumes, the plant communities tended to be increasingly limited by N (Fig. 1). Aboveground P storage significantly differed among years, but did not show a consistent temporal trend. The lowest aboveground P storage was observed in the exceptionally dry year 2005 (Table 1). Plant species richness significantly explained 14% of the variance in aboveground P storage during the study period (F1,82 = 3.1, p = 0.03). The significant effect of plant species richness remained (p b 0.03) if we fitted each functional group before plant species richness to elucidate whether the plant species richness effect was driven by the presence of particular functional groups. With time, plant species richness increasingly contributed to the explanation of the variance in aboveground P storage both for the spring and fall harvests. Again, this increase was independent of the order of plant species richness and plant functional groups in the statistical model, i.e. the increase was not driven by particular functional groups. Based on the same repeated measures model as in Table 2, in the spring harvests, aboveground P storage was increased by the presence of legumes (F1,82 = 21.5, p b 0.0001; through a positive biomass effect of legumes), and decreased by the presence of grasses (F1,82 = 13.4, p = 0.001; both through negative effects of grasses on biomass and P concentrations). If mixtures contained tall herbs, aboveground P storage was higher by a factor of 1.39 ± standard error 0.04 compared to

mixtures without tall herbs (F1,82 = 5.6, p = 0.02). The presence of small herbs had significant effects on aboveground P storage (F1,82 = 11.4, p = 0.002), but these were not consistently positive or negative. 3.2. P exploitation by the plant community Phosphorus exploitation (i.e. storage of P in aboveground biomass divided by the sum of storage of bioavailable P in soil and storage of P in aboveground biomass, Eq. [1]) was based on soil P resources that are available in the intermediate run and we assumed no changes in Table 2 Repeated measures ANOVA results of effects of plant and functional group richness and the presence or absence of functional groups on community P exploitation (P storage in aboveground biomass divided by the sum of P storage in aboveground biomass and storage of labile and moderately labile P in soil). SS refers to the Sum of Squares, dF to degrees of freedom, and F represents the F value of the corresponding factor. Significant effects are given in bold. * p b 0.05; ** p b 0.01; *** p b 0.001. Factor

SS

dF

F

Time Time * block Time * species richness Time * legumes Time * grasses Time * small herbs Time * tall herbs Block Species richness Functional group richness Legumes Species richness * legumes Grasses Species richness * grasses Small herbs Species richness * small herbs Tall herbs Species richness * tall herbs

2.56 0.50 1.54 0.36 0.17 0.05 0.09 0.38 5.64 0.38 2.67 0.27 0.53 0.43 3.53 0.61 0.88 0.12

3 10 17 3 3 3 3 3 5 3 1 4 1 4 1 4 1 4

34.16*** 2.24* 4.12*** 4.78** 2.08 0.61 1.11 0.91 8.03*** 0.90 19.02*** 0.44 3.07 0.54 27.56*** 0.99 5.27* 0.15

182

Y. Oelmann et al. / Geoderma 167-168 (2011) 178–187

a

30

b

2003

absent

P exploitation (%)

25

present

20 15 10 5 0

c

30

d

2004

P exploitation (%)

25 20 15 10 5 0

e

30

f

2005

P exploitation (%)

25 20 15 10 5 0

g

30

h

2006

P exploitation (%)

25 20 15 10 5 0

P exploitation (%)

i

30

k

2007

25 20 15 10 5 0 1

2

4

8

16

60

LE

GR

sH

tH

Plant species richness Fig. 2. Plant P exploitation depending on (a, c, e, g, i) plant species richness, and (b, d, f, h, k) the presence (black bars) or absence (white bars) of functional groups from 2003 to 2007. The line connects means of the plant species richness levels (a, c, e, g, i). Whiskers indicate the standard error (b, d, f, h, k). Significance of effects can be derived from Table 6. LE = legumes; GR = grasses; sH = small herbs; tH = tall herbs.

Y. Oelmann et al. / Geoderma 167-168 (2011) 178–187

spring periods were also significant at the 0.2 m (pb 0.001; Fig. 3), but remained a non-significant trend at the 0.1 m soil depths (Fig. 3). At our field site, pH values in soil solution ranged from 7.5 to 8.4. During the study period, Ca concentrations in soil solution were 122 ± 1 mg l −1. Assuming pH values of between 7 and 8 in soil solution result in equilibrium concentration ranges of 7 × 10 −5 to 2 × 10 -8 mg l −1 P for hydroxylapatite as the controlling mineral, 0.03 to 1.3 mg l −1 for octacalciumphosphate, and 0.5 to 4.5 mg l −1 for dicalciumphosphatedihydrate. Plant species richness did not significantly influence vwm PO4-P concentrations in any season or depth (0.06 b p b 0.6). Using the same

a

vwm PO4-P (mg l-1)

0.20

0.15

0.10

0.05

0.00

b 0.20

vwm PO4-P (mg l-1)

concentrations of the sum of labile and moderately labile P concentrations in soil during the study period. Therefore, the temporal variation in P exploitation of the mixtures paralleled that of the aboveground P storage among the five study years (Fig. 1). Phosphorus exploitation ranged from 0 (Bellis perennis L. monoculture) to 24% of the available P pool in soil. In four of the five study years, highest P exploitation was observed for legume-containing mixtures. The difference in P exploitation between low-diversity mixtures and 60-species mixtures increased with time (Fig. 2). Plant species richness explained the greatest proportion of the variance in P exploitation which again increased with time (2003: 45% explained proportion of sum of squares [SS]; 2004: not significant; 2005: 54% of SS; 2006: 63% of SS; 2007: 75% of SS). Remarkable 70 and 95% of the variance in P exploitation could be explained if plant species richness and the presence/absence of legumes, respectively, were included in the statistical model. If we reversed the order of plant species richness and functional groups in the statistical model, the effect of plant species richness remained significant (p b 0.002). However, plant species richness explained less than 29% in the reversed models. This finding was attributable to the presence of legumes or tall herbs which explained 12 to 34% and 10–16% of the variance, respectively. Although the presence of legumes contributed most to the explanation of the variance in all reversed models (2003 to 2007), plant species richness still contributed similar proportions to the explained variance as the presence of legumes. The presence of legumes in mixtures increased P exploitation (Fig. 2, Table 2). Averaged for the five study years, the effect of legumes contributed 29% to the explanation of the variance in P exploitation. Although the contribution of the effect of small herbs to the averaged variance in P exploitation was generally higher (35%) than that of the legumes, the effect of small herbs did not consistently have the same direction. There was a shift from decreased to increased P exploitation between 2005 and 2006 if small herbs were present in mixtures. The N:P ratios in aboveground biomass correlated well with our measure of P exploitation (0.25 b r b 0.52; p b 0.03). Because N:P ratios differed significantly between mixtures with and without legumes (Fig. 1b) we ran separate correlations of N:P ratios with P exploitation for mixtures with and without legumes, respectively. For legumecontaining mixtures, there were significant correlations from 2003 to 2006 (0.34 b r b 0.42; 0.005 b p b 0.03). For legume-free mixtures, the correlations between N:P ratios and P exploitation were significant in 2004 and 2006 (0.35 b r b 0.49; 0.002 b p b 0.03).

183

0.15

0.10

0.05

0.00

Total P concentrations in soil ranged from 179 to 686 mg kg −1 and were on average 479 ± standard error 10 mg kg −1. The coefficient of variation among all experimental plots was 0.19. The sum of phosphate in all fractions contributed significantly more to total P concentrations (89%) than the sum of organic P (11%). Acid extraction as a measure of stable P fractions resulted in significantly higher phosphate concentrations (HCl: 276 ± 5 and H2SO4: 100 ± 3 mg P kg −1 soil) than neutral/basic extraction (labile and moderately labile, respectively; NaHCO3: 19 ± 5 and NaOH: 38 ± 2 mg P kg −1 soil). Accordingly, stable P fractions contributed most to total P concentrations (57 ± 1% [H2SO4] and 20 ± 1% [HCl], respectively). Readily plant-available P i.e., vwm PO4-P concentrations in soil solution, ranged from 0 to 0.21 mg l−1. From spring 2003 to spring 2004, vwm PO4-P concentrations were significantly higher at the 0.1 and 0.2 m compared to the 0.3 m soil depths (based on Block 2 where all depths were equipped: 0.0001 b p b 0.01; Fig. 3). Winter periods were characterized by significantly higher vwm PO4-P concentrations than the respective subsequent spring periods at 0.3 m soil depth (0.001 b p b 0.009; Fig. 3). From spring 2004 to spring 2006, differences in vwm PO4-P concentrations between the winter and subsequent

c 0.20

vwm PO4-P (mg l-1)

3.3. Phosphorus partitioning in soil

0.15

0.10

0.05

0.00 Spring Winter Spring Winter Spring Winter Spring 2003 03/04 2004 04/05 2005 05/06 2006 Fig. 3. Temporal course of vwm PO4-P concentrations in soil solution including spring (March to May) and winter (December to February) at the (a) 0.1 m, (b) 0.2 m, and (c) 0.3 m soil depths from 2003 to 2006. The line connects means of the respective periods. Note that more than 95% of the PO4-P concentrations were below 0.04 mg l−1, i.e. below the detection limit, from winter 2006/07 to spring 2007. Therefore, we excluded these periods.

Y. Oelmann et al. / Geoderma 167-168 (2011) 178–187

statistical model as in Table 3, the presence of legumes in mixtures significantly increased vwm PO4-P concentrations at 0.2 m soil depth in winter 2003/04, winter 2004/05, spring 2005, and spring 2006 (Fig. 4) and at 0.3 m soil depth in winter 2004/05 (F1,63 = 8.9, p = 0.004; legumes present: 0.04 mg l−1; legumes absent: 0.03 mg l−1). Volumeweighted PO4-P concentrations were lower if grasses were present in a mixture at the 0.1 m soil depth in spring 2003 and at the 0.3 m soil depth in winter 2004/05 (0.002 b p b 0.02; grasses present: 0.07 and 0.02; grasses absent: 0.11 and 0.04 mg l−1, respectively). The presence of legumes decreased the contribution of labile (NaHCO3-extractable) phosphate to total P (legumes present: 3.5 ± 0.5%; legumes absent 4.5 ± 0.5%) and labile (NaHCO3-extractable) phosphate concentrations in soil (Table 3, Fig. 5). The contribution of organic P to total P and concentrations of NaHCO3-extractable P in soil were significantly lower if grasses were present in a mixture (grasses present: 2.6% ± 0.3%; grasses absent 3.2 ± 0.2%; concentrations in Table 3, Fig. 5). For the other individual P fractions and the sum of all fractions, the presence/absence of plant functional groups had neither effects on P concentrations nor on the contribution of individual fractions to total P in soil (0.0 b F b 1.2; 0.3 b p b 0.9). Plant species richness had significant effects on concentrations of moderately labile (NaOH-extractable) phosphate to total P in soil (F5,82 = 4.4; p = 0.002). Because of the large contribution of the NaOH-extractable phosphate fraction to total P the plant species richness effect was also visible for the sum of all fractions (F5,82 = 3.1; p = 0.01). However, all observed plant species richness effects resulted from significantly lower NaOH-extractable P concentrations of the 60-species mixtures compared to all other diversity levels (Tukey, 0.001 b p b 0.04). 4. Discussion 4.1. Plant species richness effects The consistent effect of plant species richness on aboveground P storage and P exploitation (Table 2, 3; Fig. 2) corresponds well with

Table 3 ANOVA results on effects of plant species and functional group richness and the presence or absence of functional groups on inorganic and organic labile (NaHCO3-extractable) phosphorus concentrations in soil. SS refers to the Sum of Squares, dF to degrees of freedom, and F represents the F value of the corresponding factor. Significant effects are given in bold. * p b 0.05; ** p b 0.01. Factor

SS

dF

NaHCO3-Pi Block Species richness Functional group richness Legumes Species richness * legumes Grasses Species richness * grasses Small herbs Species richness * small herbs Tall herbs Species richness * tall herbs

F

1.16 0.30 0.11 0.89 0.60 0.55 0.27 0.04 0.22 0.17 0.61

3 5 3 1 4 1 4 1 4 1 4

2.53 0.39 0.24 5.82* 0.98 3.33 0.41 0.20 0.32 1.04 0.93

NaHCO3-Po Block Species richness Functional group richness Legumes Species richness * legumes Grasses Species richness * grasses Small herbs Species richness * small herbs Tall herbs Species richness * tall herbs

0.55 0.97 0.71 0.43 0.08 1.82 0.35 0.20 1.54 0.05 0.65

3 5 3 1 4 1 4 1 4 1 4

0.89 0.95 1.15 2.12 0.10 10.23** 0.49 1.07 2.08 0.24 0.80

0.20 absent

vwm PO 4-P(mg l -1)

184

present

0.15

0.10

0.05

0.00

Spring 2003

Winter 03/04

Spring 2004

Winter 04/05

Spring 2005

Winter 05/06

Spring 2006

Fig. 4. Volume-weighted (vwm) PO4-P concentrations in soil solution depending on the presence (black bars) or absence (white bars) of legumes from 2003 to 2006. Whiskers indicate the standard error. Note that more than 95% of the PO4-P concentrations were below 0.04 mg l−1, i.e. below the detection limit, from winter 2006/07 to spring 2007. Therefore, we excluded these periods.

our previous findings for N at the same study site (Oelmann et al., 2007b). Highly diverse plant mixtures took up more N as indicated by increased aboveground N pools compared to less diverse mixtures. We have evidence that niche facilitation in diverse systems leads to more efficient use of N (Temperton et al., 2007) and, therefore, possibly also to increased uptake of PO4-P by plants/microbes. In accordance with this, Karanika et al. (2007) found a positive correlation between plant species richness and P storage in aboveground biomass in a pot experiment. In our study, the effect of plant species richness was consistent if the presence of legumes was accounted for in the statistical model. In such models, plant species richness contributed less but similar proportions to the explanation of the variance than the presence of legumes. Therefore, our results further stress the importance of plant species richness that is not only caused by the presence or absence of particular plant functional groups. In our study, the increased P exploitation of diverse mixtures was related to increased N:P ratios in aboveground biomass particularly in legume-containing mixtures. Therefore, shifts towards P limitation already fostered P exploitation of soil resources, although P did not seem to limit plant growth based on N:P ratios (Koerselman and Meuleman, 1996) except for few mixtures containing legumes. Because of the crucial role of legumes for N or P limitation of the plant mixtures, their abundance might govern the shift from N to P limitation in the future. There were no effects of plant species richness on either readily plant-available P concentrations in soil solution or on the more strongly bound P fractions in soil (Table 3) except for significantly lower stable P concentrations under 60-species mixtures compared to less diverse systems. Since we did not find significant differences in stable P fractions between 2002 (the year of seeding the plant mixtures) and 2007 (matched pairs t test, p N 0.2), we infer that the four 60-species mixtures were established by chance on plots with low total P concentrations. During the first 5 years, plant-available P fractions were not affected by this assumed artifact. The discrepancy between positive plant diversity effects on aboveground P storage versus no plant diversity effects on P availability in soil can be attributed to the physicochemical control of P partitioning in soil. Consequently, demand-driven mobilization mechanisms of plants such as exoenzyme exudation by plant roots or associated mycorrhiza probably control P uptake independently of the directly available P pool as measured by NaOH extraction (Hinsinger, 2001). Although we observed efficient P exploitation of diverse mixtures, it has to be kept in mind that specialists adapted to P limitation are not included in the 60 plant species used for our experiment. Therefore, plant species richness effects in established plant communities

Y. Oelmann et al. / Geoderma 167-168 (2011) 178–187

a

185

b

90

absent

NaHCO3-Pi (mg kg-1)

80

present

70 60 50 40 30 20 10 0

NaHCO3-Po (mg kg-1)

c

d

30 25 20 15 10 5 0 1

2

4

8

16

60

LE

GR

sH

tH

Plant species richness Fig. 5. Labile (NaHCO3-extractable) inorganic (a, b) and organic (c, d) P concentrations in soil in 2007 depending on (a, c) species richness, and (b, d) the presence (black bars) or absence (white bars) of functional groups. The line connects means of the plant species richness levels (a, c). Whiskers indicate the standard error (b, d). LE = legumes; GR = grasses; sH = small herbs; tH = tall herbs.

adapted to low P availability might be more pronounced than in our experimental set up. 4.2. Functional group identity effects The presence of legumes in mixtures increased aboveground P storage and thus, P exploitation (Fig. 2, Table 2). The fixation of N2 is an energy-consuming process which requires ATP, and thus increases the P demand of the legume-Rhizobium symbiosis (Aerts and Chapin, 2000; Pate, 1986). In line with the high P demand, legume-containing mixtures decreased labile Pi concentrations in soil in fall 2007 (Table 3, Fig. 5). Similarly, Hooper and Vitousek (1998) reported negative effects of legumes on resin-extractable P in soil. Asghari et al. (2005) stressed that the colonization of leguminous species by arbuscular mycorrhizal fungi further increased plant P uptake and decreased P leaching from soil. However, in our study the effect of legumes was different during cold winters and if temperatures fell below 0 °C in spring (Table 1, Fig. 4). During these periods, vwm PO4-P concentrations in soil solution under legume-containing mixtures were higher than in mixtures without legumes. Legumes are frequently used as cover crops in agriculture to improve N (and P) availability in soil for subsequent crops (Ledgard, 2001; Ranells and Wagger, 1997). In this context, the increased risk of P leaching from decaying legumes/legume nodules during freeze-thaw cycles was already stressed (Ulen et al., 2005). Both, potentially increased P leaching during cold periods and increased P removal with the harvest could lead to an increasing scarcity of P in legume-containing mixtures in the long run. Therefore, positive effects of legumes on biomass production and aboveground N storage found in the Jena Experiment during the first 5 years after establishment (Marquard et al., 2009) might only be possible at the expense of P availability

in soil. Considering the threshold value suggested by Koerselman and Meuleman (1996), up to now, plant growth in most legumecontaining mixtures is not limited by P. However, in the long run, the contribution of legumes to positive plant species richness effects might change if the grassland systems shift to P limitation. For instance, a strong decline in biomass production of legumes from 2006 to 2008 and N2 fixation rates correlating negatively with leaf P concentrations suggested a P limitation of legumes with increasing age of the biodiversity experiment (Roscher et al., 2011). Despite pronounced effects of grasses on N cycling reported for our study site (Oelmann et al., 2007b) we hardly found significant effects of grasses on P cycling. The presence of grasses in mixtures decreased labile Porg concentrations in soil (Table 3, Fig. 5). Particularly grasscontaining mixtures associated with an extensive rooting system (Bessler et al., 2009) might exude exoenzymes that enable P mobilization from organic compounds (Hinsiger, 2001). The extensive rooting system again might ensure efficient subsequent P uptake (Hooda et al., 1999). In accordance, the presence of grasses in mixtures decreased vwm PO4-P concentrations in soil solution. For our study site, an excessive root length density of 40 cm root length per cm³ was reported (Bessler et al., 2009) which might be partly attributable to the strong physicochemical control of P concentrations in soil solution and the resulting low P availability. The presence of small and tall herbs mainly affected aboveground P storage and P exploitation (Table 2), however, not consistently. We speculate that the tall herbs used in our experiment benefited from the initially elevated P concentrations because of fertilizer remains. In contrast, small herbs might have suffered from the initially low N supply. As the P concentration decreased with increasing time after establishment of the experiment, the tall herbs were no longer able to substantially contribute to aboveground P storage and exploitation.

186

Y. Oelmann et al. / Geoderma 167-168 (2011) 178–187

In contrast, the small herbs benefited from increasing N concentrations because of the accumulation of organic matter in response to the conversion of the study site from arable to grassland use (Habekost, 2008). Generally, all functional groups contributed most to the explanation of the variance in P exploitation in the driest year 2005. This finding highlights the importance of diversity in functional traits under extreme conditions which are thought to guarantee temporal stability in diverse ecosystems (Cardinale et al., 2007; Tilman et al., 2006; van Ruijven and Berendse, 2009). 4.3. Temporal course of plant diversity effects From spring 2003 to spring 2004, vwm PO4-P concentrations in soil solution at the 0.1 and 0.2 m soil depths were higher than in the subsequent seasons (Fig. 3). Fertilizer-derived P-containing minerals have higher solubility products than natural P-containing minerals which would result in higher dissolved PO4-P concentrations in soil solution (Hart et al., 2004). Because of the low solubility product of natural P-containing minerals, dissolved PO4-P concentrations will be low (b0.01 mg l −1) if dissolution of natural P-containing minerals is the driving mechanism. The observed vwm PO4-P concentrations in soil solution were initially within the calculated ranges of fertilizer-derived phosphates and decreased below calculated concentrations of fertilizer-derived phosphates in winter 2004/05 (Fig. 3). The decrease in dissolved PO4-P concentrations during our monitoring period suggests that this pool was continuously depleted. The fact that more than 95% of PO4-P concentrations were below the detection limit from winter 2006/07 to spring 2007 highlights the importance of the dissolution equilibrium of natural P-containing minerals towards the end of our study period. Despite this obvious physicochemical control induced by the composition of P-containing minerals that is not expected to change throughout the year, we observed higher vwm PO4-P concentrations in soil solution during winter than during spring periods (Fig. 3). Biomass production reaches maximum values in spring before the first harvest at the end of May. Therefore, in spring the P demand of the plants for growth is higher. Mechanisms of plants to explore P resources in soil e.g., through symbiosis with mycorrhizal fungi (Hinsinger, 2001) seem to have thus been efficient enough to reduce vwm PO4-P concentrations in soil solution in spring. The consistent correlations between N:P ratios of aboveground biomass and P exploitation of the plant community and between P concentrations in aboveground biomass and biomass production and the increasing slopes of the regression lines with time stress the relevance of P for productivity. Furthermore, plant species richness increasingly contributed to the explanation of P exploitation with time (Table 2). Therefore, the importance of the P mobilization strategies of plants (including symbiosis with fungi) might even further increase given the fact that fertilization ceased before the experiment and N deposition still remains high (Oelmann et al., 2007a). 5. Conclusions i. Positive plant species richness effects on aboveground P storage and P exploitation were consistently significant and did not depend on particular plant functional groups. In contrast, plant species richness did not influence readily plant-available P in soil solution or P fractions in soil. ii. The increased P demand of legumes for N2 fixation resulted in increased P exploitation by legume-containing mixtures. On the other hand, increased readily plant-available P concentrations in soil solution during cold periods indicate an increased risk of P leaching under legume-containing mixtures. Both effects are related with increased P losses by biomass removal with the harvest and enhanced P leaching and thus, suggest that productivity in legume-containing mixtures might be

limited by P in the near future. The effect of the other functional groups on P cycling was not consistent during the five study years. However, the pronounced effect of all functional groups on P exploitation in the driest study year 2005 highlights the importance of plant functional diversity as a guarantee for stability of ecosystem processes in diverse grasslands. iii. Positive plant species richness effects aboveground P storage and P exploitation increased across time again independent of the presence of plant functional groups. Therefore, plant strategies to efficiently acquire P from soil seemed to gain importance during the experimental course which might be further promoted by continuously high N deposition and a possible shift to P limitation of productivity in the future.

Acknowledgements We gratefully acknowledge the contribution of the co-initiators of the Jena Experiment, E.-D. Schulze and B. Schmid. Thanks to G. Gleixner and S. Steinbeiß, Max Planck Institute for Biogeochemistry, Jena, for their substantial input into this work. Special thanks to Agnes Rehmus for her great commitment in running the CFA. We thank the many people who helped with the management of the experiment especially the gardeners S. Eismann, S. Hengelhaupt, S. Junghans, U. Köber, K. Kuntze, H. Scheffler and U. Wehmeier. Thanks also to all the helpers during the weeding campaigns. The Jena Experiment is funded by the Deutsche Forschungsgemeinschaft (DFG, Wi 1601/4-1,-2, FOR 456), with additional support from the Friedrich Schiller University of Jena and the Max Planck Society, and the Swiss National Science Foundation. References Aerts, R., Chapin, F.S., 2000. The mineral nutrition of wild plants revisited: a reevaluation of processes and pattern. Advances in Ecological Research 30, 1–67. Alt, F., Oelmann, Y., Harold, N., Schrumpf, M., Wilcke, W., 2011. Phosphorus partitioning in grassland and forest soils of Germany as related to land-use type, management intensity, and land use-related pH. Journal of Plant Nutrition and Soil Science 174, 195–209. Asghari, H.R., Chittleborough, D.J., Smith, F.A., Smith, S.E., 2005. Influence of arbuscular mycorrhizal (AM) symbiosis on phosphorus leaching through soil cores. Plant and Soil 275, 181–193. Bardgett, R.D., Streeter, T.C., Bol, R., 2003. Soil microbes compete effectively with plants for organic-nitrogen inputs to temperate grasslands. Ecology 84, 1277–1287. Bessler, H., Temperton, V.M., Roscher, C., Buchmann, N., Schmid, B., Schulze, E.-D., Weisser, W.W., Engels, C., 2009. Aboveground overyielding in grassland mixtures is associated with reduced biomass partitioning to belowground organs. Ecology 90, 1520–1530. Brundrett, M.C., 2002. Coevolution of roots and mycorrhizas of land plants. New Phytologist 154, 275–304. Cardinale, B.J., Wright, J.P., Cadotte, M.W., Carroll, I.T., Hector, A., Srivastava, D.S., Loreau, M., Weis, J.J., 2007. Impacts of plant diversity on biomass production increase through time because of species complementarity. Proceedings of the National Academy of Sciences of the United States of America 104, 18123–18128. Ciais, P., Reichstein, M., Viovy, N., Granier, A., Ogee, J., Allard, V., Aubinet, M., Buchmann, N., Bernhofer, C., Carrara, A., Chevallier, F., De Noblet, N., Friend, A.D., Friedlingstein, P., Grunwald, T., Heinesch, B., Keronen, P., Knohl, A., Krinner, G., Loustau, D., Manca, G., Matteucci, G., Miglietta, F., Ourcival, J.M., Papale, D., Pilegaard, K., Rambal, S., Seufert, G., Soussana, J.F., Sanz, M.J., Schulze, E.D., Vesala, T., Valentini, R., 2005. Europe-wide reduction in primary productivity caused by the heat and drought in 2003. Nature 437, 529–533. Critchley, C.N.R., Chambers, B.J., Fowbert, J.A., Sanderson, R.A., Bhogal, A., Rose, S.C., 2002. Association between lowland grassland plant communities and soil properties. Biological Conservation 105, 199–215. Cross, A.F., Schlesinger, W.H., 1995. A literature review and evaluation of the Hedley fractionation: applications to the biogeochemical cycle of soil phosphorus in natural ecosystems. Geoderma 64, 197–214. Eisenhauer, N., Bessler, H., Engels, C., Gleixner, G., Habekost, M., Milcu, A., Partsch, S., Sabais, A.C.W., Scherber, C., Steinbeiss, S., Weigelt, A., Weisser, W.W., Scheu, S., 2010. Plant diversity effects on soil microorganisms support the singular hypothesis. Ecology 91, 485–496. Ellenberg, H., 1996. Die Vegetation Mitteleuropas mit den Alpen in ökologischer, dynamischer und historischer Sicht, 5. edn. Ulmer, Stuttgart. 1095 pp. Galloway, J.N., Dentener, F.J., Capone, D.G., Boyer, E.W., Howarth, R.W., Seitzinger, S.P., Asner, G.P., Cleveland, C.C., Green, P.A., Holland, E.A., Karl, D.M., Michaels, A.F., Porter, J.H., Townsend, A.R., Vörösmarty, C.J., 2004. Nitrogen cycles: past, present, and future. Biogeochemistry 70, 153–226.

Y. Oelmann et al. / Geoderma 167-168 (2011) 178–187 Gilbert, J., Gowing, D., Wallace, H., 2009. Available soil phosphorus in semi-natural grasslands: assessment methods and community tolerances. Biological Conservation 142, 1074–1083. Habekost, M., 2008. Influence of plant diversity on soil organic carbon storage and microbial transformation of organic carbon in soil. Dissertation. University of Jena, Germany. Hart, M.R., Quin, B.F., Nguyen, M.L., 2004. Phosphorus runoff from agricultural land and direct fertilizer effects: a review. Journal of Environmental Quality 33, 1954–1972. Hedley, M.J., Stewart, J.W.B., Chauhan, B.S., 1982. Changes in inorganic and organic soil phosphorus fractions induced by cultivation practices and laboratory incubations. Soil Science Society of America Journal 46, 970–976. Hinsinger, P., 2001. Bioavailability of soil inorganic P in the rhizosphere as affected by root-induced chemical changes: a review. Plant and Soil 237, 173–195. Hooda, P.S., Moynagh, M., Svoboda, I.F., Edwards, A.C., Anderson, H.A., Sym, G., 1999. Phosphorus loss in drainflow from intensively managed grassland soils. Journal of Environmental Quality 28, 1235–1242. Hooper, D.U., Vitousek, P.M., 1998. Effects of plant composition and diversity on nutrient cycling. Ecological Monographs 68, 121–149. Hooper, D.U., Chapin, F.S., Ewel, J.J., Hector, A., Inchausti, P., Lavorel, S., Lawton, J.H., Lodge, D.M., Loreau, M., Naeem, S., Schmid, B., Setala, H., Symstad, A.J., Vandermeer, J., Wardle, D.A., 2005. Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecological Monographs 75, 3–35. Janssens, F., Peeters, A., Tallowin, J.R.B., Bakker, J.P., Bekker, R.M., Fillat, F., Oomes, M.J.M., 1998. Relationship between soil chemical factors and grassland diversity. Plant and Soil 202, 69–78. Karanika, E.D., Alifragis, D.A., Mamolos, A.P., Veresoglou, V., 2007. Differentiation between responses of primary productivity and phosphorus exploitation to species richness. Plant and Soil 297, 69–81. Kluge, G., Müller-Westermeier, V., 2000. Das Klima ausgewählter Orte der Bundesrepublik Deutschland: Jena. Berichte des Deutschen Wetterdienstes, p. 213. Koerselman, W., Meuleman, V., 1996. The vegetation N:P ratio: a new tool to detect the nature of nutrient limitation. Journal of Applied Ecology 33, 1441–1450. Kuo, S., 1996. Phosphorus. In: Sparks, D.L., Page, V., Helmke, P.A., Loeppert, R.H., Soltanpour, P.N., Tabatabai, M.A., Johnston, C.T., Sumner, M.E. (Eds.), Methods of Soil Analysis — Part 3 Chemical Methods: Soil Sci. Soc. Am. , pp. 869–920. Madison, Wisconsin, USA. Ledgard, S.F., 2001. Nitrogen cycling in low input legume-based agriculture, with emphasis on legume/grass pastures. Plant and Soil 228, 43–59. Marquard, E., Weigelt, A., Temperton, V.M., Roscher, C., Schumacher, J., Buchmann, N., Fischer, M., Weisser, W.W., Schmid, B., 2009. Plant species richness and functional composition drive overyielding in a 6-year grassland experiment. Ecology 90, 3290–3302. McCrea, A.R., Trueman, I.C., Fullen, M.A., Atkinson, M.D., Besenyei, L., 2001. Relationships between soil characteristics and species richness in two botanically heterogeneous created meadows in the urban English West Midlands. Biological Conservation 97, 171–180. McKane, R.B., Johnson, L.C., Shaver, G.R., Nadelhoffer, K.J., Rastetter, E.B., Fry, B., Giblin, A.E., Kielland, K., Kwiatkowski, B.L., Laundre, J.A., Murray, G., 2002. Resource-based niches provide a basis for plant species diversity and dominance in arctic tundra. Nature 415, 68–71. Murphy, J., Riley, J.P., 1962. A modified single solution method for the determination of phosphate in natural waters. Analytica Chimica Acta 27, 31–36. Negassa, W., Leinweber, P., 2009. How does the Hedley sequential phosphorus fractionation reflect impacts of land use and management on soil phosphorus: a review. Journal of Plant Nutrition and Soil Science 172, 305–325. Niklaus, P.A., Kandeler, E., Leadley, P.W., Schmid, B., Tscherko, D., Körner, C., 2001. A link between plant diversity, elevated CO2 and soil nitrate. Oecologia 127, 540–548. Oelmann, Y., Kreutziger, Y., Temperton, V.M., Buchmann, N., Roscher, C., Schumacher, J., Schulze, E.-D., Weisser, W.W., Wilcke, W., 2007a. Nitrogen and phosphorus budgets in experimental grasslands of variable diversity. Journal of Environmental Quality 36, 396–407. Oelmann, Y., Wilcke, W., Temperton, V.M., Buchmann, N., Roscher, C., Schumacher, J., Schulze, E.-D., Weisser, W.W., 2007b. Soil and plant nitrogen pools as related to

187

plant diversity in an experimental grassland. Soil Science Society of America Journal 71, 720–729. Oelmann, Y., Buchmann, N., Gleixner, G., Habekost, M., Roscher, C., Rosenkranz, S., Schulze, E.-D., Steinbeiss, S., Temperton, V.M., Weigelt, A., Weisser, W.W., Wilcke, W., 2011. Plant diversity effects on above- and belowground N pools in temperate grassland ecosystems: development in the first five years after establishment. Global Biogeochemical Cycles 25, GB2014. Pate, J.S., 1986. Economy of symbiotic N fixation. In: Givnish, T.J. (Ed.), On the Economy of Plant Form and Function. Cambridge Univ Press, Cambridge, pp. 299–325. Ranells, N.N., Wagger, M.G., 1997. Grass-legume bicultures as winter annual cover crops. Agronomy Journal 89, 659–665. Roscher, C., Schumacher, J., Baade, J., Wilcke, W., Gleixner, G., Weisser, W.W., Schmid, B., Schulze, E.-D., 2004. The role of biodiversity for element cycling and trophic interactions: an experimental approach in a grassland community. Basic and Applied Ecology 5, 107–121. Roscher, C., Thein, S., Weigelt, A., Temperton, V.M., Buchmann, N., Schulze, E.-D., 2011. N2 fixation and performance of 12 legume species in a 6-year grassland biodiversity experiment. Plant and Soil 341, 333–348. Scherer-Lorenzen, M., Palmborg, C., Prinz, A., Schulze, E.-D., 2003. The role of plant diversity and composition for nitrate leaching in grasslands. Ecology 84, 1539–1552. Schmid, B., Joshi, J., Schläpfer, F., 2002. Empirical evidence for biodiversity-ecosystem functioning relationships. In: Kinzig, A., Tilman, D., Pacala, S. (Eds.), Functional Consequences of Biodiversity: Experimental Progress and Theoretical Extensions. Monographs in Population Biology 33. Princeton University Press, Princeton, pp. 120–150. Spehn, E.M., Hector, A., Joshi, J., Scherer-Lorenzen, M., Schmid, B., Bazeley-White, E., Beierkuhnlein, C., Caldeira, M.C., Diemer, M., Dimitrakopoulos, P.G., Finn, J.A., Freitas, H., Giller, P.S., Good, J., Harris, R., Hogberg, P., Huss-Danell, K., Jumpponen, A., Koricheva, J., Leadley, P.W., Loreau, M., Minns, A., Mulder, C.P.H., O'Donovan, G., Otway, S.J., Palmborg, C., Pereira, J.S., Pfisterer, A.B., Prinz, A., Read, D.J., Schulze, E.-D., Siamantziouras, A.S.D., Terry, A.C., Troumbis, A.Y., Woodward, F.I., Yachi, S., Lawton, J.H., 2005. Ecosystem effects of biodiversity manipulations in European grasslands. Ecological Monographs 75, 37–63. Styles, D., Coxon, C., 2007. Meteorological and management influences on seasonal variation in phosphorus fractions extracted from soils in western Ireland. Geoderma 142, 152–164. Temperton, V.M., Mwangi, P.N., Scherer-Lorenzen, M., Schmid, B., Buchmann, N., 2007. Positive interactions between nitrogen-fixing legumes and four different neighbouring species in a biodiversity experiment. Oecologia 151, 190–205. Tilman, D., Wedin, D., Knops, J., 1996. Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature 379, 718–720. Tilman, D., Reich, P.B., Knops, J.M.H., 2006. Biodiversity and ecosystem stability in a decade-long grassland experiment. Nature 441, 629–632. Trenbath, R.B., 1974. Biomass production of mixtures. Advances in Agronomy 26, 177–210. Ulen, B., Aronsson, H., Torstensson, G., Mattsson, L., 2005. Phosphorus and nitrogen turnover and risk of waterborne phosphorus emissions in crop rotations on a clay soil in southwest Sweden. Soil Use and Management 21, 221–230. van Ruijven, J., Berendse, F., 2003. Positive effects of plant species diversity on productivity in the absence of legumes. Ecology Letters 6, 170–175. van Ruijven, J., Berendse, F., 2009. Long-term persistence of a positive plant diversity– productivity relationship in the absence of legumes. Oikos 118, 101–106. Vitousek, P.M., Howarth, R.W., 1991. Nitrogen limitation on land and in the sea — how can it occur. Biogeochemistry 13, 87–115. Vitousek, P.M., Porder, S., Houlton, B.Z., Chadwick, O.A., 2010. Terrestrial phosphorus limitation: mechanisms, implications, and nitrogen–phosphorus interactions. Ecological Applications 20, 5–15. von Felten, S., Hector, A., Buchmann, N., Niklaus, P.A., Schmid, B., Scherer-Lorenzen, M., 2009. Belowground nitrogen partitioning in experimental grassland plant communities of varying species richness. Ecology 90, 1389–1399. Wassen, M.J., Venterink, H.O., Lapshina, E.D., Tanneberger, F., 2005. Endangered plants persist under phosphorus limitation. Nature 437, 547–550.